Nonlinear stability control of autonomous vehicles: a MIMO D2-IBC solution

Stability control represents one of the most challenging tasks in autonomous vehicle dynamics. Within this context, model uncertainty may play a crucial role in determining the quality of the overall control system. In this work, we deal with nonlinear vehicle stability control design by using the D2-IBC (Data Driven - Inversion Based Control) method, wherein the dynamics of the system are modeled from data not to optimize the open-loop model matching but to maximize the closed-loop performance. Specically, a multivariable extension of the method is derived and discussed in detail as far as the stability control application is concerned. This method will prove its eectiveness on a multi-body simulator of a 4-wheel steering autonomous vehicle.